File: insertps-from-constantpool.ll

package info (click to toggle)
llvm-toolchain-20 1%3A20.1.6-1~exp1
  • links: PTS, VCS
  • area: main
  • in suites: experimental
  • size: 2,111,304 kB
  • sloc: cpp: 7,438,677; ansic: 1,393,822; asm: 1,012,926; python: 241,650; f90: 86,635; objc: 75,479; lisp: 42,144; pascal: 17,286; sh: 10,027; ml: 5,082; perl: 4,730; awk: 3,523; makefile: 3,349; javascript: 2,251; xml: 892; fortran: 672
file content (21 lines) | stat: -rw-r--r-- 1,067 bytes parent folder | download | duplicates (9) 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
 
; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py UTC_ARGS: --no_x86_scrub_mem_shuffle --version 4
; RUN: llc -mtriple=i686-unknown-unknown -mattr=+sse4.1 < %s | FileCheck %s --check-prefix=X86
; RUN: llc -mtriple=x86_64-unknown-unknown -mattr=+sse4.1 < %s | FileCheck %s --check-prefix=X64

; Test for case where insertps folds the load of an insertion element from a constant pool.

define <4 x float> @fold_from_constantpool(<4 x float> %a) {
; X86-LABEL: fold_from_constantpool:
; X86:       # %bb.0:
; X86-NEXT:    insertps $0, {{\.?LCPI[0-9]+_[0-9]+}}+4, %xmm0 # xmm0 = mem[0],xmm0[1,2,3]
; X86-NEXT:    retl
;
; X64-LABEL: fold_from_constantpool:
; X64:       # %bb.0:
; X64-NEXT:    insertps $0, {{\.?LCPI[0-9]+_[0-9]+}}+4(%rip), %xmm0 # xmm0 = mem[0],xmm0[1,2,3]
; X64-NEXT:    retq
  %1 = call <4 x float> @llvm.x86.sse41.insertps(<4 x float> %a, <4 x float> <float 0.0, float 1.0, float 0.0, float 0.0>, i8 64)
  ret <4 x float> %1
}

declare <4 x float> @llvm.x86.sse41.insertps(<4 x float>, <4 x float>, i8) nounwind readnone